Artificial intelligence fusion for predicting survival of rectal cancer patients using immunohistochemical expression of Ras homolog family member B in biopsy

Abstract

Aim: The process of biomarker discovery is being accelerated with the application of artificial intelligence (AI), including machine learning. Biomarkers of diseases are useful because they are indicators of pathogenesis or measures of responses to therapeutic treatments, and therefore, play a key role in new drug development. Proteins are among the candidates for biomarkers of rectal cancer, which need to be explored using state-of-the-art AI to be utilized for prediction, prognosis, and therapeutic treatment. This paper aims to investigate the predictive power of Ras homolog family member B (RhoB) protein in rectal cancer. Methods: This study introduces the integration of pretrained convolutional neural networks and support vector machines (SVMs) for classifying biopsy samples of immunohistochemical expression of protein RhoB in rectal-cancer patients to validate its biologic measure in biopsy. Features of the immunohistochemical expression images were extracted by the pretrained networks and used for binary classification by the SVMs into two groups of less and more than 5-year survival rates. Results: The fusion of neural search architecture network (NASNet)-Large for deep-layer feature extraction and classifier using SVMs provided the best average classification performance with a total accuracy = 85%, prediction of survival rate of more than 5 years = 90%, and prediction of survival rate of less than 5 years = 75%. Conclusions: The finding obtained from the use of AI reported in this study suggest that RhoB expression on rectal-cancer biopsy can be potentially used as a biomarker for predicting survival outcomes in rectal-cancer patients, which can be informative for clinical decision making if the patient would be recommended for preoperative therapy.

Wavelet Scattering Networks of Deep-Learning Features for Robust Discovery of Protein Markers in Rectal Cancer

“…In comparison with studies reported in 18,19 , not only the AI-based approach developed in current work was tested with many types of the cancer tissues (biopsy, tumor, metastasis, adjacent normal) with RT and non-RT, but also able to consistently achieve much higher prediction rates in all cases. In comparison between using the pretrained CNN and SVM models for classification, training of the SVM model with the double convolution-based extracted features could provide a much faster computational speed than the pretrained CNNs.…”

“…Many efforts have been attempted to discover biomarkers in cancers [7][8][9][10][11][12][13] . More specifically, to include a few reports, mutations of BRAF, KRAS, and p53 genes were identified to cause the development of CRC 14 , proteins of guggulsterone, which is a plant phytosteroid, were found to have associations with growth inhibitory effects in human CRC cells 15 ; APRIL, BAFF, IL8, and MMP2 expressions were involved in the inflammatory CRC tumor microenvironment 16 ; multiple biomarkers in young CRC patients 17 ; DNp73 18 and RhoB 19 expressions on immunohistochemistry (IHC) of rectal cancer tumor and biopsy samples; faecal microbial biomarkers can be used for early diagnosis of CRC 20 ; and gut microbiome (stool, blood, tissue, bowel fluid) have been reviewed as main sample sources of biomarkers for screening CRC 21 .…”

Wavelet scattering networks in deep learning for discovering protein markers in a cohort of Swedish rectal cancer patients

“…The reported results can be reproduced using the IHC imaging data and MATLAB codes implemented in this paper. Both data and Matlab codes are available at the author's personal homepage: https://sites.google.com/view/tuan-d-pham/codes under the titles “Double convolutional learning of protein expression in rectal cancer” for the Matlab codes, “Artificial intelligence‐based 5‐year‐survival prediction and prognosis of DNp73 expression in rectal cancer patients” for DNp73 data, 18 and “Rectal cancer biopsy” for RhoB data 19 …”

“…com/ view/ tuan-d-pham/ codes under the titles "Double convolutional learning of protein expression in rectal cancer" for the Matlab codes, "Artificial intelligence-based 5-yearsurvival prediction and prognosis of DNp73 expression in rectal cancer patients" for DNp73 data, 18 and "Rectal cancer biopsy" for RhoB data. 19…”

“…5,6 Numerous endeavors have been made to identify biomarkers associated with various aspects of cancer. [7][8][9][10][11][12][13] To cite a few examples: mutations in the BRAF, KRAS, and p53 genes have been implicated in CRC development 14 ; the protein guggulsterone, a plant phytosteroid, has shown connections to inhibiting the growth of human CRC cells 15 ; the expression of APRIL, BAFF, IL8, and MMP2 has been linked to the inflammatory microenvironment of CRC tumors 16 ; multiple biomarkers have been explored in young CRC patients 17 ; the evaluation of DNp73 18 and RhoB 19 expressions through immunohistochemistry (IHC) in rectal cancer tumor and biopsy samples; fecal microbial biomarkers have demonstrated potential for early CRC diagnosis 20 ; and the gut microbiome, encompassing stool, blood, tissue, and bowel fluid samples, has been extensively investigated as a primary source of biomarkers for CRC screening. 21 Being a novel dimension for biomedical research and health care, artificial intelligence (AI) and particularly convolutional neural networks (CNNs) known as deep learning are increasingly making themselves indispensable to the reality of precision or personalized medicine.…”